基于改进YOLOv5s的道路坑洼检测算法  

Road Pothole Detection Algorithm Based on Improved YOLOv5s

作　　者：白芮 徐杨 王彬 张雯雯 BAI Rui;XU Yang;WANG Bin;ZHANG Wen-wen(College of Big Data&Information Engineering,Guizhou University,Guiyang 550025,China;Guiyang Aluminum-magnesium Design and Research Institute Co.LTD,Guiyang 550025,China)

机构地区：[1]贵州大学大数据与信息工程学院,贵州贵阳550025 [2]贵阳铝美设计研究院有限公司,贵州贵阳550025

出　　处：《计算机与现代化》2023年第6期69-75,共7页Computer and Modernization

基　　金：贵州省科技计划项目(黔科合支撑[2021]一般176)。

摘　　要：针对现有目标检测算法难以对道路坑洼进行精准检测、检测速度慢等问题,提出一种基于改进YOLOv5s的道路坑洼检测算法。首先在YOLOv5s主干网络中融入坐标注意力(Coordinate Attention,CA)模块,使模型不仅捕获跨通道信息,还捕获方向和位置敏感信息,有助于模型更准确地定位和识别检测对象;然后在空间金字塔池化(Spatial Pyramid Pool,SPP)模块中采用软池化SoftPool改进最大池化操作,保留更详细的特征信息;在特征融合阶段,使用基于内容的功能重组(Content-Aware ReAssembly of Features,CARAFE)对多尺度特征融合中上采样进行改进,动态生成自适应内核,可以在一个大的感受野内聚集上下文信息;最后,使用Alpha-IoU对损失函数进行改进,提高边框回归精度。实验结果表明,改进的YOLOv5s算法在平均精度上较原始网络提高了4.6个百分点,与其他主流算法SSD、Faster R-CNN、YO⁃LOv3、YOLOv3-tiny、YOLOv4-tiny相比检测精度有较大提升。Aiming at the problem that existing target detection algorithms are difficult to accurately detect road potholes and the detection speed is slow,a road pothole detection algorithm based on improved YOLOv5s is proposed.Firstly,CA(Coordinate at⁃tention)module is integrated into YOLOv5s backbone network,so that the model can capture not only cross-channel informa⁃tion,but also direction perception and position sensitive information,which is helpful for the model to locate and identify the de⁃tected object more accurately.Then,SoftPool is adopted in Spatial Pyramid Pool(SPP)module to improve the maximum pooling operation and retain more detailed characteristic information.In the feature fusion stage,Content-Aware ReAssembly of FEatures(CARAFE)is used to improve the up-sampling of multi-scale feature fusion and dynamically generate an adaptive kernel,which can gather context information in a large receptive field.Finally,Alpha-IoU is used to improve the loss function and improve the margin regression accuracy.Experimental results show that the average accuracy of the improved YOLOv5s algorithm is 4.6 per⁃centage points higher than that of the original network,and the detection accuracy of the improved YOLOv5s algorithm is greatly improved compared with other mainstream algorithms such as SSD,Faster R-CNN,YOLOv3,YOLOv3-tiny and YOLOv4-tiny.

关 键 词：深度学习 坑洼检测 坐标注意力 最大池化 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]